kasw'08 - invited talk

Lehrstuhl Informatik V(Informationssysteme)

Prof. Dr. M. JarkeI5-RK-0808-1

CUELC

Ralf KlammaRWTH Aachen University

KASW Workshop, I-Media, September 3, 2008

Community-Oriented Knowledge Acquisition and Analysis



CUELC

Agenda

Media & Knowledge Communities Case Studies

– Disturbances– Scientific Communities– Community Measures

Conclusions & Outlook Agency & Patienthood in Digital Networks

Agency & Patienthood in Digital Networks



CUELC

Learning & Knowledge ManagementIndividual / Community Perspective

[Nonaka & Takeuchi, 1995]

[Ullman, 2004]

Semantic Knowledgesemiotic concepts

documentation

Verbal

wordslinguistic data

Non-verbal

image, icon, indexvideo blogs, diagrams, images, photographies

Episodic Knowledge memory of experiencing past episodes

web blogs, narratives

Declarative Knowledge Procedural Knowledgesensomotoric skills, procedural scripts

non-documented routines and operations



CUELC

Semiotics in the Tradition of Ferdinand de Saussure (1957 - 1913)

comprehension / articulationactivation of community information system

human neural networkArtifacts of community information system

in p

rese

ntia

in a

bsen

tia

performanceParole

competenceLangue



CUELC

Hypotheses

1. A semiotic knowledge system is dynamic and it changes every time it is activated.

2. The meaning of a concept is determined by how it interacts with other concepts and by how it can be distinguished from other concepts in the knowledge system. (Positive and Negative Knowledge)

3. The knowledge system is carried by a material medium. The modality of the medium influences knowledge structures.



CUELC

Cross-MediaTheory of transcription

Pre-“texts“

TranscriptCross-Media Transcription

Understandand Criticize

Jäger, Stanitzek: Transkribieren - Medien/Lektüre 2002

Strategies of transcriptivity Collection of learning materials are re-structured by new media Design is specific for media and communities by default

Strategies of addressing Social Software promotes the globalization of address spaces Personalization and adaptive strategies are mission critical for CoPs

Strategies of localisation Re-organization of local practices is stimulated by new media like Social Software Need to model practice explicitly



CUELC

Babylonian Talmud: A very old Hypertext

• Scroll/book/printed book • Talmud schools (Jeshiwot) • Authoritative knowledge source• Dialogic encyclopedia • Structure of complex texts• Connected knowledge

Transcribe? Address? Localize?



CUELC

CESE: Multi-lingual Cross-Media System

Published in: DS-NELL 2000, ICALT 2002, ICWL 2002, WWW 2003



CUELC

Research Approach:Reflective Learning Network

Collaborative adaptive learning network

Mining tools for Communities

Measure, Analyse, Simulate

Social Software

Development

Assessment requirements for Communities

Support evolving learning communities (repeated assessment of community requirements)

Based on Preece 2001, cf. I-KNOW 2006 for details



CUELC

Solution idea for Reflective Support:Cross-Media Social Network Analysis

Interdisciplinary multidimensional model of digital networks– Social network analysis (SNA) is defining measures for social

relations– Actor network theory (ANT) is connecting human and media agents– I* framework is defining strategic goals and dependencies– Theory of media transcriptions is studying cross-media knowledge

social softwareWiki, Blog, Podcast, IM, Chat, Email, Newsgroup, Chat …

i*-Dependencies(Structural, Cross-media)

Members(Social Network Analysis: Centrality,

Efficiency)

network of artifactsMicrocontent, Blog entry, Message, Burst, Thread,

Comment, Conversation, Feedback (Rating)

network of members

Communities of practice

Media Networks



CUELC

Simplified Meta Model for ANT using Latour

Actor

Member NetworkLearningService

Medium Artifact

Attribute has

stores creates is affected by belongs go

represents consumes performs ranks

… MatchRetrievalBrowse Search

isA

isA

Latour: On Recalling ANT, 1999Klamma, Spaniol, Cao: A model for social software, IJKL 2007



CUELC

Modeling dependencies using the i* framework

Eric S. K. Yu, Towards Modeling and Reasoning Support for Early-Phase Requirements Engineering, RE 1997

Network

Coordinator

Gatekeeper

Hub

Member

Iterant Broker

URL

isA

isA

isA

isA

Coordination

Artifact

Communication

Legend:

AgentGoalResource Task



CUELC

Disturbances in Cross-media Social Networks

What is a disturbance?– Sensing an incompatibility

between theories exposed and theories-in-use

Disturbances are starting points of learning processes– Disturbances disturb,

prevent … but they are creating reflection

Disturbances are hard to detect or to forecast



CUELC

Pattern Language for PALADIN: Example Troll

Troll Pattern: This pattern tries to discover the cases when a troll exists in a digital social network. A troll in the network is considered a disturbance.

Disturbance: (EXISTS [medium | medium.affordance = threadArtefact]) &

(EXISTS [troll |(EXISTS [thread | (thread.author = troll) & (COUNT [message | (message.author = troll) & (message.posted = thread)]) > minPosts]) & (~EXISTS[ thread1, message1| (thread1.author1 != troll) &

(message1.author = troll & message1.posted = thread1 ]))])])

Forces: medium; troll; network; member; thread; message; url

Force Relations: neighbour(troll, member); own thread(troll, thread)

Solution: No attention must be paid to the discussions started by the troll. Rationale: The troll needs attention to continue its activities. If no attention is paid, he/she

will stop participating in the discussions. Pattern Relations: Associates Spammer pattern.



CUELC

Pattern Discovery ProcessPattern

Disturbance

Variables

Pattern Template

Disturbance

VariablesPattern Parameters

Pattern Template Instance

Pattern Instance

Disturbance

Variables Pattern Parameters

Forces ForceRelations

Rationale

Dependencies

Description Solution

Pattern Relations

Disturbance Instances

Variables Pattern Parameters

Digital Social Network

1. Set pattern parameters

2. Instantiate disturbances

3. Evaluate disturbances

4a. Change Pattern Parameters

4b. Apply Pattern Solution



CUELC

PALADIN Case Study

10 patterns of disturbance over 119 social network instances, 17359 individuals, 215 345 mails

Pattern Occurrences Remarks

Burst 22 The pattern finds out topics which were very important for certain period of time. Scalability is necessary.

No Conversationalist 76 The existence implies little communication in the network.

No Questioner 67 The existence implies that the network is not popular.

No Answering Person

61 Occurs in small networks. The effects of the lack of an answering person must be further checked with content analysis.

Troll 2 Troll occurs very rarely in cultural communities. True negatives exist.

Spammer 86 Spammers can be found often in discussion groups. False positives exist.

Leader 37 The pattern occurs in the network centered around a member.

No Leader 40 Occurs in big networks where the members are distributed in different clusters.

Structural Hole 67 Occurs for members having neighbors with only one contact.

Independent Discussions

13 Occurs in large networks where disconnected subnetworks exist. Scalability is necessary.



CUELC

Impact of research community on individuals

Academic event modeling- Unstructured data of academic events - Diversity of additional media: Photos, Videos, Blogs, Wikis…=> A model for academic events and their communities

documentation

Events recommendation tool for researchers- Design a community based recommendation algorithm

Events communities analysis and visualization.- Community analysis from community of practice point of view



CUELC

AERCS: Evolution of Scientific Communities



CUELC

Evolution of community

VLDB 1990 VLDB 1995

VLDB 2000 VLDB 2006



CUELC

Community visualization – ACM SIGMOD example

ACM SIGMOD



CUELC

Models of Community Success

Reference Model: D&M IS Success Model (1992) – Based on >100 Empirical/Conceptual Studies – Validated by Independent Studies Updated

Model: Integration of Current Concepts – Mobility (Mobile Context) – Multimedia Communities



CUELC

Success Classification & Measurement

Quantitative: Monitoring User-Service – Communication Logging – Mobile Context Information – MobSOS Monitoring Module

Quantitative & Qualitative: Survey – Online User Surveys (Questionnaire) – MobSOS Survey Module

Subjective Objective

Quantitative Monitoring Survey

Qualitative Survey



CUELC

MobSOS Monitoring Module

Client: Capture & Transmit Mobile Context Information

Testbed: Log Communication & Mobile Context



CUELC

Mobile Service Oracle for Success

Monitoring of service invocations Time and position tracking of a service call Recognition of patterns in user behaviour



CUELC

Storytelling Expertfinding

New Measure for Knowledge in a Community

Expert value

Mean: 0,2624# Entries: 99.778

Freq

uenc

y



CUELC

Story-tellling Expert Finding

KeywordsExpert values

Knowledge Value of Community sorted by keywords#

Reco

mm

enda

tions

Expert

Amateur



CUELC

Conclusions Media and Communities shape knowledge structures

– Semiotic systems depend on media– Communities set goals and means

Case Studies– Social Patterns in Communities– Evolution of Scientific Communities– Community Success Models– Expert finding in Communities

Further research– Uncertainty in Tagging systems– Continuous elicitation of community needs– Emotional dimension of collective intelligence

kasw'08 - invited talk

Technology